GaryGaulin
Posts: 5385
Joined: Oct. 2012
|
Ah ha!
Since it takes 1 cycle for the spin sensor to sense being in a spin its being 1 cycle back made that sensor barely useful anymore, and I'm apparently seeing it making the best of what little sense that leaves it with (when it was seeing nothing at all). Making the Spin sensor act without delay improved things, no longer favored one lobe, by leaving the spinning direction more to the other. But that reduced how well it did with optimum timing, which is the fastest learner of them all, and did better with delayed Spin that takes a full cycle of motor torque (one lobe timestep followed by other's timestep) to be considered spinning.
All did well staying fed, within a few pixels of each other at almost 100% full at all times. It helps to be at optimal hold time (less than 1/2 cycle for sensory to reach memory address in and write) but still works with a long delay.
I'm not sure how that might apply to motor neuron delays, but that's what happens in this model, from not being optimal speed with memory.
--------------
The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.
|
